The clinical research community has adopted the use of preprint servers, which provide outlets for preliminary reports of research that has not been peer-reviewed.1 Preprint servers support open scholarship, allow research to be disseminated quickly, offer opportunities for peer feedback before formal submission to a journal, and have been increasingly adopted by the biological, physical, and economic scientific communities.2-4 However, for preprint potential to be realized in clinical research, peer-reviewed journals must be willing to consider manuscripts that were previously posted on preprint servers (preprints) for publication. Because systematic information about contemporary clinical journal policies on preprints is lacking,5 our objective was to assess the preprint publication policies of the 100 clinical journals with the highest impact factors.
This cross-sectional study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. For this study, we used InCites Journal Citation Reports (JCR) to identify journals across all fields with a 2018 journal impact factor greater than 5. We manually screened all identified journals by title and categories on JCR to find the 100 top-ranked clinical journals and included only those that publish original research.
For each qualifying journal, we checked a succession of resources to ascertain its editorial policy on preprints: the journal website; the publisher website; the Transpose Database6; and the first 10 pages of a Google search containing the journal name and the term preprint. Once a preprint policy was located, we classified each journal’s policy by the following categories: preprints allowed (if preprints will be considered for publication), case-by-case determination (if preprints are evaluated on an individual basis), and preprints prohibited (if preprints will not be considered for publication). Data were collected on April 23, 2020. We conducted descriptive analyses using Microsoft Excel (Microsoft Corporation).
Among the 100 top-ranked clinical journals, the median (interquartile range) impact factor was 13.7 (10.7-19.0). Most journals (86 [86%]) allowed preprints (Table 1). In contrast, 13 journals (13%) evaluated each preprint independently to determine whether to reject it on the basis of its prior preprint status (case-by-case determination). Only 1 journal (1%) had a policy that prohibited preprints (preprints prohibited). There was no association between the median impact factor and the category of preprint policy (Table 2).